Researchers at Johns Hopkins University have developed a new method using machine learning to visualize the strength of synapses, the connectors that enable communication between nerve cells in the brain. The new technology allows for better comprehension of how learning, aging, injury, and disease affect these connections in the human brain. Synapses can fluctuate over time, and the ability to track these changes in living animals could lead to a better understanding of brain function. Because of the high density and small size of synapses, it has been difficult for scientists to visualize the shifting chemistry of synaptic messaging. By employing machine learning, the researchers were able to use noisy imaging data to recover the signal sections they needed to observe.
The technique used genetically modified mice with synapses that fluoresced green when exposed to light, enabling the researchers to track individual synapses in living animals and repeatedly capture photos of the same synapses over time. The method could provide insight into the synaptic alterations that occur in different illness and injury scenarios, with one potential use being the study of synaptic changes in animal models of Alzheimer’s disease. The research was conducted by a team of scientists from a range of fields, including molecular biology and artificial intelligence, in cooperation at the multidisciplinary Kavli Neuroscience Discovery Institute.
